Weihua FANG

Academy of Disaster Reduction and Emergency Management, (MoCA & MOE of China)

Beijing Normal University

27 February 2014, Chengdu, China

Integrate Science, Technology and Financeinto the Coordination on

Regional Disaster Governance

Outline1. Background

2. Concepts

3. Case StudyMulti-hazard Database and Risk MappingCatastrophe Risk Modeling and Risk FinanceOther Risk Assessments in China

4. Discussions

1. Background: Complex Disaster System

1. Background: Regional Impacts

Trans-boundary Hazards and Direct LossEarthquake, TsunamiTyphoon, FloodSand Storm, ……

Catastrophic DisastersBeyond local/national coping capacity

Trans-boundary Indirect Impacts EconomicEcological Environmental

Regional Coordination and Collaboration

2. Concepts: from Risk Management to Governance

2. Concepts: Risk Governance Framework

2. Concepts: Stakeholders of Risk Governance

2. Concepts: Disaster Management Cycle

2. Concepts: Disaster Management Cycle

What stage is the most concerned by regional organizations and why?

What stage is the most concerned by regional organizations and why?

2. Concepts: Disaster Management Cycle

2. Concepts: from Emergency Response to Risk Governance

What kind of capacities should be built?How to take proactive measures?

What are the roles of science and technology?

Spatial and Temporal Heterogeneity Where? How often? How Strong?

Policy-Making Target Users: National/Province/County Govs. What-if info:

Casualty Building Damage Economic Lose Evacuation Population

Public DRR Practice Education ……

Others

3.1 Case I: Purpose

Hazards Earthquake Flood Typhoon Drought Snow Storm Sand Storm Storm Surge Landslide Hail

Frost Forest Fire Grassland Fire Chemical incidents ……

Exposure Data Population

County/township/zip-code 1km*1km

GDP County/township/zip-code 1km*1km

Building Year Story Type Occupancy …

Infrastructure Transportation Utility Evacuation site Hospital

Crops Wheat, Corn Rice…..

Loss Data MoCA Statistics

1949-2009 County-level Province level Hazard-specific

Insurance Data Policy Claim

Case Study Data Earthquakes Flood Typhoon Drought Wildfire ……

Satellite-based Wildfire Drought Earthquake Flood………

3.1 Case I: Database

Auxiliary Dataset GIS, social-economic Coping capacity……

Map Resolution 1km grid County

Map Types Quantitative Semi-quantitative Categories

3.1 Case I: Mapping Methods

3.1 Risk Mapping: Earthquake

3.1 Risk Mapping: Earthquake

3.1 Risk Mapping: Flood

3.1 Risk Mapping: Typhoon (Wind)

3.1 Risk Mapping: Typhoon (rainfall)

3.1 Risk Mapping: Typhoon (economic loss)

3.1 Risk Mapping: Storm Surge (ranking)

3.1 Risk Mapping: Drought (wheat)

3.1 Risk Mapping: Drought (corn)

3.1 Risk Mapping: Landslide

3.1 Risk Mapping: Landslide

3.1 Risk Mapping: Snowstorm

3.1 Risk Mapping: Hail (ranking)

3.1 Risk Mapping: Frost (ranking)

3.1 Risk Mapping: Forest Fire

3.1 Risk Mapping: Grassland Fire

3.1 Risk Mapping: Grassland Fire

3.1 Risk Mapping: Insurance Policy and Claim

3.1 Risk Mapping: Insurance Policy and Claim

3.1 Risk Mapping: Integration

Stochastic Event Module: Track and Intensity ModelingHazard Module: Wind and Rainfall ModelingVulnerability Module: Linking Hazard and LossRisk Module: Statistics, Actuary, Cost-Benefit Analysis

3.2 Case 2: Components of Typhoon Risk Model

Stochastic event set (right, 620 years) based on historical tracks (left, 62 years, 1949-2010): West-Northern Pacific

Genesis, Moving, Landing, Decay (filling), Lysis

3.2 Case 2: Stochastic Event Set Generation

What •Typhoon wind field model is used to estimate the spatial and temporal distribution of typhoon wind.

Why•The historical observation data is inadequate in space and time with limited observation year range

How•Parametric Model•Numerical Model

3.2 Case 2: Parametric Wind Model

Boundary Layer Model: Estimation of the surface mean wind speed adjusted from gradient mean wind speedKey Input: Surface roughness length, determined from LULC data.

3.2 Case 2: Parametric Wind Model

Directional Topographic Effect

(ERN-Capra, 2013)

3.2 Case 2: Parametric Wind Model

10 20 30 40 50 60 70 80 90 100

5.5

6.0

6.5

7.0

7.5

8.0

8.5

� � (min)

��

(m/s)

10min mean wind speed 1min mean wind speed

0

0

0

,,

VG T

TTV

Example: Set τ=3 s, T0=10min, get G3,600

Gust Factor

Gust Factor Definition:

3.2 Case 2: Parametric Wind Model

Maximum Sustained Wind (10min) Maximum Sustained Wind after roughness modification

Maximum Sustained Wind after roughness, and topographic modification

Gust Wind (3s) after modification of roughness, topographic gust factor

3.2 Case 2: Parametric Wind Model

Modeling of instantaneouswind field to wind swath

Output and Verification

Model verification using observation data

3.2 Case 2: Parametric Wind Model

TC key parametersIntensity (MWS，Pmin）Position (lon, lat)Translating speed and direction

Underlying surface conditionstopographic condition(DEM, slope aspect, etc.)SSTland-sea distribution

Environmental variable and general circulationVertical Wind ShearMoisture and water vapor transport westerly trougheasterly wave

FY-2C 1-hour PRE rainfall rate at 2009-09-16 14:00UTC

Conceptual Model of Typhoon Rainfall Structure

3.2 Case 2: Parametric Rainfall Model

Wind Load & Resistance: Example of Rural Residential Building in Coastal Area of China

3.2 Case 2: Building Vulnerability Model

Empirical Vulnerability Curve: Example of Rubber Tree to Wind in Hainan Island

Totally Destroyed Moderate DamageServe Damage

3.2 Case 2: Rubber Tree Vulnerability Model

Output of Loss Probability Model

1. Annual Exceedance Probability (AEP)

2. Occurrence Exceedance Probability (OEP)

3. Exceeding Probability Curve (EP)

4. Fine-resolution Risk Mapping (30m /1000m)

5. Risk of Insured Property (Deductibles & Limits)

6. Portfolio Management

3.2 Case 2: Loss Probability Modeling

Loss Distribution of an Example Farm

Loss Events Cumulated Distribution Probability of Loss

3.1 Case 2: Loss Probability Modeling

，，

Annual Aggregate Loss of Rubber Tree(Pure Insurance Rate)

3.1 Case 2: Insurance Rate Calculation

3.1 Case 2: Insurance Portfolio

Cat Model can Help Understand the Risks of Complicated Portfolio

3.1 Case 2: Payouts Triggered by Wind Speed

Benefits of Parametric Insurance No moral hazard. No adverse selection Lower operating costs Transparency No cross-subsidization Immediate disbursement. Reinsurance and securitization.

Stochastic Event and Wind Field Model Basis risk Model bias Technical limitations of insurable hazards Education

51

3.2 Case 2: Parametric Typhoon Insurance

A Parametric Insurance Project (Research and Pilot)Supported by Ministry of Finance of China

Applications in Insurance Industry Index-based Wind Risk Insurance of Rubber Tree in Hainan Province

(World Bank Project 2013) County-level Reference Insurance Rate by CIRC Supporting Multi-peril Property Insurance of PICC

Stochastic Event and Wind Field Model Stochastic event sets + wind field model + numerical storm surge model

(ADCIRC) Mapping coastal flood hazard (flooding areas of various return periods) Land Use Planning

Synthetic tracks + ADCIRC mapping of Probable Maximum Storm Surge (PMSS) CBDM Evacuation Planning

Wind field model + numerical wave model (SWAN) Wave Risk

Stochastic Event and Rain Field Model Stochastic event sets + wind field model + runoff model mapping riverine

flood risk

3.2 Case 2: Many Application Potentials

Cross-Platform: Windows, *NIX, Mac DB & GIS: PostGIS Model library: Java Desktop System: Java Cloud (B/S): user only need provide exposure data

Development Plan (3 products) CycloneRisk CycloneWarning (proto-type) CycloneLoss

3.3 Case 2: Welcome to Join OpenCyclone!

Ministry of Civil Affairs Multi-hazards, focusing on loss

Ministry of Water Resource, Ministry of Agriculture Floods, Droughts

China Earthquake Administration Earthquakes

Ministry of Land Resource Geological Disasters

China Marine Administration Storm Surge, Wave, Tsunami, Sea Ice, Sea Level Rise

Community-Level Risk Assessment Contingency Planning Evacuation

3.3 Other Risk Assessments

4. Discussions

1. Disaster Mitigation Mainstreaming and Planning Cost-benefit Analysis / Budget Application Priority Analysis

2. Regional Risk Finance Regional Catastrophe Fund?

1. Regional Emergency Response Regional Emergency Response Fund?

Thank you for your attention

The end.

Weihua FANGweihua.fang@bnu.edu.cn

•Ph.D., Associate ProfessorAcademy of Disaster Reduction and Emergency Management, MoCA & MOE, China

Beijing Normal University